Recessionary changes at work and employee well-being: The protective roles of national and workplace institutions

Abstract

The recession following the 2008 financial crisis brought major changes to employees’ experiences at work. We investigate the adverse effects of two of such changes: perceived organizational distress and job deterioration. We also examine the extent to which institutions at national level (employment protection legislation and collective bargaining coverage) and at workplace level (employment contract and union membership) may act as buffers against these effects. Using data from 21 European countries, we show that recessionary changes were associated with reduced psychological well-being and greater levels of work–nonwork interference.

Keywords

Europe institutions job deterioration perceived organizational distress recession well-being

Introduction

The 2008 financial crisis caused a decline in profitability and growth, and many organizations were forced to adjust working arrangements, reduce employment or working hours and rely on more flexible or low-cost employment practices (Shimer, 2010). Consequently, employees faced greater job insecurity (Iverson and Zatzick, 2011), higher probability of unemployment (Datta et al., 2010) and deteriorating employment conditions (Wood and Ogbonnaya, 2018). In the year immediately following the onset of the crisis, employment in the EU fell by approximately 4.3 million, while unemployment among those aged below 25 years reached almost 20 percent by 2009, and there was a noticeable decline in labour market participation (Heyes, 2011).

There has been considerable research into the macroeconomic and organizational consequences of the recession that followed this crisis (Cingano, et al., 2010; Datta et al., 2010; Lane and Milesi-Ferretti, 2010), but less attention has been given to employees’ experiences and responses (Burgard and Kalousova, 2015). Limited research on how workers fared during this period presents a significant gap in our understanding of recessionary effects on employee well-being. We contribute to existing knowledge by examining employee perceptions of recessionary changes at work and the consequences for their well-being. We use the term ‘perceived recessionary changes’ in reference to perceptions of adverse employment events, such as organizational distress (financial shocks and workforce reduction) and job deterioration (JD) such as reduction in pay and job insecurity, which are directly or indirectly associated with economic recession.

We link the adverse well-being effects of recessionary changes to important institutional factors designed to buffer against such effects (Appelbaum, 2011). These include national labour market institutions aimed at protecting workers from unemployment and reduced employment standards (employment protection laws and collective bargaining), as well as workplace-level arrangements that improve workers’ job security and protect them against reduced job quality (secure employment contracts and union representation). This focus is particularly useful when governments have questioned the efficacy of existing institutional arrangements (Carr and Chung, 2014), when policymakers are paying closer attention to health and well-being as indicators of national and societal success (Green et al., 2016) and when unions and collective bargaining have been in decline (Valizade et al., 2016).

We investigate our hypotheses using data from the 2010 European Social Survey (ESS), matched to country-level data from the OECD measures of national employment institutions. The timing of the data collection allows us to operationalize our research within the context of the 2008 recession.

Recessionary changes and employee well-being

Economic recessions are typically associated with higher inflation, intensified product market competition and economic distress for organizations (Datta et al., 2010). Employers are often forced to re-evaluate their business and employment strategies by making changes to employment benefits and wages, restructuring jobs and adjusting the quality of work responsibilities (Curl and Kearns, 2015; Griffith and Macartney, 2014). While such actions may stabilize organizational operations, they expose workers to greater employment insecurity, reductions in real employment benefits and constrained opportunities for career development (Felstead, et al., 2012; Warren, 2015), all of which contribute to poor well-being (Green et al., 2016; Wood and Ogbonnaya, 2018).

A number of studies have shown how economic recession may affect different groups of workers. In the financial sector, for example, Snorradóttir et al. (2015) found that workers who remained in their company following downsizing experienced a greater reduction in psychological well-being than those who were made redundant. Giorgi et al. (2015) found evidence in private and public sector organizations of higher work-related stress and job dissatisfaction, particularly among workers who reported strong fears about the potential destructive effects of economic crisis. Similarly, Curl and Kearns (2015) found that individuals in 15 deprived communities who were affected by recessionary events fared worse than their counterparts who were not affected, mainly because of limited household income and reduced levels of social support.

To extend this work, we concentrate on employees’ direct reactions to perceived employment changes associated with recession. We distinguish between two aspects: the organizational context (perceived organizational distress, POD) and the job context (job deterioration, JD). The concept of organizational distress captures the organization-level consequence of adverse external shocks during recessionary periods (Giorgi et al., 2015), which may involve increased financial difficulties and significant workforce reductions. The notion of JD measures the extent to which important characteristics that shape job quality have deteriorated over time, irrespective of the institutional or organizational context (Green et al., 2013); this may involve pay cuts, less interesting work or reduced working hours, and job insecurity. The distinction between POD and JD is crucial for a number of reasons. First, it shapes our understanding of how broader organization-level phenomena might induce a different pattern of outcomes compared to employees’ direct experiences at the job level (Jiang et al., 2013). Second, organizational and job-specific sources of employment insecurity are likely to have varying effects on different aspects of employee well-being (Sverke and Hellgren, 2002). Third, labour market institutions are expected to influence organizational and job-specific sources of employment insecurity in different ways (Carr and Chung, 2014).

We aim to understand how POD and JD may affect two outcomes: psychological well-being and work–nonwork interference. We follow prior research (Wood and Michaelides, 2016) that makes a clear distinction between the likely consequences of work-related stressors on both individuals’ perceptions of life more generally and their perceptions about the interface between work and family. We define psychological well-being as individuals’ subjective assessments of different emotional states in relation to the quality of their lives (Budd and Spencer, 2015). Psychological well-being may be measured by positive emotions such as cheerfulness and enthusiasm and the relative absence of negative emotions such as anxiety. Work–nonwork interference is defined in terms of the spillover of affective, cognitive or physical states from work into other spheres of life, resulting in competing demands on an individual’s time and resources (Wood and Michaelides, 2016). We prefer this term to ‘work-family conflict’, as individuals without family responsibilities may also experience some form of interference in the work and nonwork domains of life (Wood et al., 2018).

A number of complementary theoretical perspectives provide the rationale for linking POD and JD to both psychological well-being and work–nonwork interference. Psychological contract theory, for example, regards employees and employers as capable of developing a set of assumed mutual obligations over time. In periods of recession, when the employer is likely impose pay cuts or reductions in working hours, employees may perceive the employer as reneging on these obligations, and this may lead to a breakdown in employment relationships (Wood and Ogbonnaya, 2018). In a distinct but similar vein, uncertainty management theory suggests that recessionary events create greater levels of insecurity that heightens perceptions of injustice or unfairness among employees (van den Bos and Lind, 2002). Similarly, job stress theory (Shoss, 2017) may explain the possible adverse effects of unfavourable work conditions on employee well-being. Accordingly, unfavourable work conditions are organizational stressors that induce feelings of employment pressure and uncertainty. They also create anxieties that undermine workers’ sense of control over their work situation (Shoss, 2017; Sverke and Hellgren, 2002). Our first hypothesis is therefore derived from these perspectives: we expect recessionary events that adversely affect employees’ work conditions will negatively influence their health and well-being.

H1: Experience of POD and JD is negatively associated with employees’ psychological well-being, and positively associated with work–nonwork interference.

Moderating roles of national and workplace institutions

To understand the circumstances under which perceived recessionary changes may influence employee well-being, we investigate the moderating roles of important institutional factors, understood as regulations, procedures, frameworks or arrangements that shape both socio-economic and employment behaviours (Donnellanet al., 2012). Well-functioning institutions allow people to deal with the likely adverse consequences of a recession (Groot et al., 2011); they could, for example, provide employees with a sense of work-related stability, safeguard their experience of job security and consequently improve their well-being (Burgard and Kalousova, 2015; Reeves et al., 2014). We concentrate on two national institutions, employment protection legislation (EPL) and collective bargaining coverage, and two workplace institutions, type of employment contract and union membership.

EPL protects workers from unfair treatment by the employer and defines procedures that employers must follow when making decisions about working arrangements, hiring, firing or redundancy. However, research evidence regarding the benefits of such laws is mixed. For example, high firing costs associated with employment protection may increase employers’ reluctance to hire new employees or invest in enhanced human capital (Groot et al., 2011). Strict EPL may also increase rates (and duration) of unemployment and undermine organizational flexibility. Conversely, other research suggests that EPL may act as a buffer against adverse health and well-being outcomes during difficult times at work (Burgard and Kalousova, 2015) by increasing workers’ bargaining power over threats to job losses and making redundancies more difficult (Heyes, 2011). Workers may be shielded from feelings of powerlessness, work conflict and occupational changes that disrupt relationship networks in the workplace (Carr and Chung, 2014). We might expect therefore that the negative effects of unfavourable employment (or recessionary) conditions on employee well-being will be weaker where EPL offers more protection to workers.

H2: The negative impact of POD and JD on well-being is weaker for employees in countries with higher employment protection than countries with lower employment protection.

Unions and collective bargaining arrangements represent a critical means of protecting workers from the adverse effects of economic and labour market shocks (Rubery and Rafferty, 2013). Unions protect employee interests by deploying a range of strategies to influence employment standards through collective bargaining and to represent members in workplace decision-making (Bryson et al., 2013). By providing a collective voice for employees during workplace negotiations, union representation can offset the power imbalance in employment relationships and increase employees’ sense of empowerment (Furåker and Bengtsson, 2013). At the national level, for instance, high bargaining coverage strengthens workers’ capacity to influence the terms of employment and improve the conditions under which they work; it also creates a context in which workers can negotiate more favourable job-related benefits (Flavin et al., 2010).

There is some evidence that union-based employee representation can enhance employment security among workers during recessionary periods (Bryson and Freeman, 2013; Campos Lima and Artiles, 2011), by inhibiting employers from imposing austerity plans unilaterally in the face of economic crises (Cascio, 2005). It follows that workers in a national context where collective agreements are more widespread may be confident that their interests are adequately represented, and the impact of recessionary events on their well-being will therefore be weaker.

H3: The negative impact of POD and JD on well-being is weaker for employees in countries with high bargaining coverage than where coverage is lower.

As with national-level institutions, institutional factors at the workplace level may also moderate the relationship between perceived recessionary changes and employee well-being. We focus on two important workplace institutions, type of employment contract and union representation.

Employment may be governed by permanent, temporary or short-term contracts. Depending on the quality of employment contract, employees may experience work-related anxieties and other negative emotions if the future existence of their job is perceived as under threat (Sverke and Hellgren, 2002). For example, those with more permanent employment agreements may not experience a heightened sense of insecurity when their employing organization faces recessionary events or external shocks. Prior research has also shown that temporary workers are more likely to report a higher level of perceived insecurity than permanent workers in adverse economic situations (e.g. Rigotti et al., 2015). De Cuyper and De Witte’s (2007) study, however, deviates from this line of evidence. Based on a sample of 477 workers in Belgium, De Cuyper and De Witte (2007) found evidence that the negative effects of perceived job insecurity on well-being were considerably weaker for temporary workers than permanent workers. This apparent paradox is generally explained as a product of the differing psychological contracts that different employment statuses are likely to give rise to: temporary workers have lower expectations that their employer will protect their employment during a period of economic distress than permanent workers do (Cheng and Chan, 2008). Thus, we might expect the likely adverse consequences of recessionary events on well-being to be stronger for workers employed under a more permanent employment contract.

H4: The negative impact of POD and JD on well-being is stronger for workers employed on a permanent employment contract than for those with a non-permanent contract.

Analysis of the American General Social Survey by Brochu and Morin (2012) showed union members as likely to feel insecure about future job prospects, particularly during periods of economic recession. They outlined industry factors rather than self-selection (implying that members self-select into union jobs) as possible explanations for their findings. In a recent study, Douglas et al. (2017) found evidence that union membership mitigates adverse well-being outcomes among workers who feel less secure about their jobs. They explained that union membership provides support and emotional resources for employees to manage the prospects of losing important aspects of their job, and this in turn minimizes emotional exhaustion and cynicism. This argument reflects the underlying premise of our next hypothesis: that union membership provides workers with a sense of protection against adverse recessionary conditions, notwithstanding the heightened sense of insecurity that those conditions might generate.

H5: The negative impact of POD and JD on well-being is weaker among union members than among non-unionists.

Methodology

ESS data were matched with 2010 OECD measures of EPL and collective bargaining coverage (OECD, 2016). We test our hypotheses using data from two independent sources: the 2010 ESS (2016) for measures of POD, JD, employee well-being and workplace-level institutional moderators; and the OECD database for national institutional moderators. The ESS is designed to monitor public attitudes, values, beliefs and behaviours across European countries; data are collected by random sampling involving hour-long face-to-face interviews. The 2010 ESS covers 28 countries, but because of missing data our study is limited to 21 of these. We focus on respondents who were employees in paid employment at the time of data collection (N = 16,271) and exclude those who were either self-employed or worked for their own or family businesses. The median number of workers in sampled countries is 730 (range: 532–1325). . Details of items used for each of our variables, and summary statistics are provided in Table A1 of the Online Appendix.

Measures

POD was measured by two items from the ESS recessionary impact module, which reports employees’ reflections about the previous 3 years, the period in which most European countries experienced the recession. The first item asked respondents how far their work organization had experienced financial difficulty in the past 3 years (1 = ‘no financial difficulty’ to 4 = ‘a great deal of financial difficulty’). The second relates to workforce reduction: respondents were asked whether the number of employees at the organization had increased or decreased in the past 3 years (1 = ‘increased a lot’ to 5 = ‘decreased a lot’).

JD was also derived from the 2010 ESS recessionary impact module. The four items for this variable asked respondents whether each of the following happened to their jobs in the past 3 years: had to do less interesting work, had to take a pay cut, had to work shorter hours and/or had less security in their job. Each was measured on a binary scale (0 = ‘no’ or 1 = ‘yes’).

Psychological well-being was measured by three items in which respondents were asked how often they had felt the following emotions in the past 2 weeks: ‘cheerful and in good spirits’; ‘calm and relaxed’ and ‘active and vigorous’ (1 = ‘at no time’ to 6 = ‘all of the time’).

Work–nonwork interference was measured by three items in which respondents were asked how often they ‘keep worrying about work problems when they are not working’; ‘feel too tired after work to enjoy the things they would like to do at home’; and ‘find that their job prevents them from giving their time to their partners or family’ (1 = ‘always’ to 5 = ‘never’).

Employment contract was measured by a single item that required respondents to state whether their main job involved an unlimited work contract, a limited work contract or no written contract. This item was coded as 0 = ‘non-permanent contract (contract of limited duration or no written contract)’ or 1 = ‘permanent contract (contract of unlimited duration)’.

Union membership was measured by a single item that asked whether respondents had ever been a member of a trade union or similar organization. This was coded as 0 = ‘non-union member’ or 1 = ‘union member’.

EPL was measured by the 2010 OECD index for individual and collective dismissals. This index is based on information from existing statutory laws and collective agreements that deal with the procedures and costs involved in dismissing or hiring individuals or groups of workers.

Collective bargaining coverage was derived from the 2010 OECD index for the national percentage of employees with the right to bargain. This index is calculated in terms of the proportion of employees covered by collective agreements, divided by all wage earners with bargaining rights within the country.

Peculiar features of our nested data preclude us from applying the OECD indices directly to our analysis. For example, the ratio of national units (the 21 ESS countries) to individual units (16,271 employees) provides insufficient statistical power to achieve unbiased standard errors and accurate parameter estimates (Scherbaum and Ferreter, 2009: 351). Moreover, data from the OECD indices are not normally distributed and the intervals between values vary considerably. Given these constraints, we recoded each institutional factor as a grouping variable to reflect two possible institutional contexts: 0 = ‘low institutional context’ or 1 = ‘high institutional context’. We used the median value of each institutional factor as the cut-off point (see details in Table A2 of the Online Appendix). For example, EPL was coded as ‘0’ where the OECD index was less than or equal to 2.22 and ‘1’ where it was greater. Similarly, bargaining coverage was coded as ‘0’ where the corresponding index was less than or equal to 64 percent, and ‘1’ where it was greater.

We included a number of control variables by following precedents in previous studies (Carr and Chung, 2014; Houdmont et al., 2012; Reeves et al., 2014; Wood and Ogbonnaya, 2018:): disability, gender, age, education level, organization size, contracted weekly hours, sector and number of years in paid employment. All control variables were derived from the 2010 ESS.

Analytical procedure

Data were analysed by structural equation modelling (SEM) with latent constructs. We used the Mplus software program (Version 7.1) together with the robust maximum likelihood estimator that adjusts for non-normality of data (Muthén and Muthén, 2010). We applied two weighted estimates in the analyses to ensure more accurate results: first, design weights to account for unequal probability of sample selection given the sample designs adopted by participating ESS countries; second, population size weights to correct for the different population sizes of each participating ESS country. We also used the Mplus complex survey data syntax to correct for potential sources of error through non-independence of data (respondents nested within different countries). This syntax adjusts standard errors for clustering effects and is used when hypotheses are made at the individual level of analysis.

The measurement component of SEM was first estimated to ensure that all measurement items (excluding the four institutional factors, which are categorical items) are discrete and reliable indicators of their respective latent constructs. This model yielded adequate goodness-of-fit: X² = 326.282; df = 48; p < 0.001; RMSEA = 0.02; CFI = 0.94; TLI = 0.92; SRMR = 0.02. All factor loadings were statistically significant and greater than 0.40 (see Table A1 of the Online Appendix).

We tested H1 by regressing the latent constructs of psychological well-being and work–nonwork interference on the latent constructs of POD and JD. Control variables were incorporated into the covariance matrix of the regression equations.

Hypotheses 2–5 refer to the moderating effects of two national-level institutions (EPL and bargaining coverage), and corresponding institutions at workplace level (employment contract and union membership). We assessed these moderating effects by multiple group analysis, which belongs to a family of SEM procedures known as mean and covariance structures (MACS) analyses (Byrne, 2012). Moderated effects in multiple group analysis are estimated by specifying the moderator as a grouping variable; allowing parameter estimates (such as regression slopes) to vary across categories of the grouping variable, and assessing whether parameter estimates are statistically different across categories of the grouping variable. Using this approach, we examined four multiple group moderation models, each testing whether the direct impact of POD and JD on psychological well-being and work–nonwork interference, respectively, were statistically different across the two categories of each institutional factor. Wald Chi-square test of parameter equalities was applied where necessary to determine the statistical difference in regression estimates (Muthén and Muthén, 2010). If the Wald test p value is < 0.05, there is evidence that estimates are statistically different.

Results

Table 1 shows that the bivariate correlations among POD, JD, psychological well-being and work–nonwork interference are in the hypothesized direction. POD is negatively correlated with psychological well-being and positively correlated with work–nonwork interference. JD also has a negative correlation with psychological well-being and a positive correlation with work–nonwork interference. The correlation between POD and JD is positive, and that between psychological well-being and work–nonwork interference is negative.

Table 1.

Bivariate correlations among study variables.

Variables		1	2	3
1	POD	–
2	JD	0.25***	–
3	Psychological well-being	−0.09***	−0.13***
4	Work–nonwork interference	0.09***	0.14***	−0.31***

Sample size (N): 16,271 in 21 countries.

***

p < 0.001.

Table 2 shows standardized regression coefficients for the direct impact of POD and JD on employee well-being. POD is not significantly associated with psychological well-being (β = −0.07, p > 0.05), although it has a positive relationship with work–nonwork interference (β = 0.10, p < 0.01). JD is however significantly associated with both well-being outcomes – negatively with psychological well-being (β = −0.11, p < 0.05) and positively with work–nonwork interference (β = 0.15, p < 0.001). With the exception of a non-significant relationship between POD and psychological well-being, our data found support for H1: that perceived recessionary changes are likely to induce negative consequences on employee well-being.

Table 2.

Direct effects of POD and JD on workers’ well-being.

	Psychological wellbeing (β)	Work–nonwork interference (β)
POD	−0.07	0.10**
JD	−0.11*	0.15***
Control variables
Disability	−0.18***	0.12***
Gender (male)	0.08***	−0.07**
Education level	0.02*	0.08***
Age (ref. 40–49)
17 and below	0.02***	−0.02
18–21	0.04	−0.02
22–29	0.08***	−0.03
30–39	0.03**	0.02
50–59	−0.02	−0.02
60–64	−0.00	−0.04***
65 and above	0.02*	−0.06***
Organization size	0.00	0.02*
Contracted weekly hours	−0.01	0.19***
Sector (private)	−0.02	0.04*
Years in paid employment	0.05	0.05

Sample size (N): 16,271 in 21 countries.

All regression coefficients (β) are standardized estimates.

p < 0.05.

p < 0.01.

***

p < 0.001.

Table 3 shows the moderating effects of national and workplace institutions, derived by multiple group analysis. We provide standardized regression coefficients for each moderated effect, and where applicable, we report corresponding Wald test estimates.

Table 3.

Moderating effects of national- and workplace-level institutional factors.

EPL	Institutional context (β)		Wald test
	Low	High	Wald test
POD → well-being	−0.02	−0.11	No statistical difference, p = 0.18
POD → interference	0.14***	0.08	Not required
JD → well-being	−0.17***	−0.06	Not required
JD → interference	0.21***	0.13**	No statistical difference, p = 0.12
Bargaining coverage	Institutional context (β)
	Low	High	Wald test
POD → well-being	−0.04	−0.13*	Not required
POD → interference	0.11***	0.10**	No statistical difference, p = 0.82
JD → well-being	−0.15***	−0.05	Not required
JD → interference	0.21***	0.07**	Statistically different, p = 0.00
Contract	Non-permanent (β)	Permanent (β)	Wald test
POD → well-being	−0.08	−0.09*	Not required
POD → interference	0.11**	0.10**	No statistical difference, p = 0.69
JD → well-being	−0.09	−0.11**	Not required
JD → interference	0.18***	0.15***	No statistical difference, p = 0.83
Union membership	No (β)	Yes (β)	Wald test
POD → well-being	−0.09	−0.10	No statistical difference, p = 0.89
POD → interference	0.09*	0.18**	No statistical difference, p = 0.18
JD → well-being	−0.10*	−0.07	Not required
JD → interference	0.17***	0.08	Not required

Sample size (N): 16,271.

Wald test not required where statistical difference exists.

All regression coefficients (β) are standardized estimates.

p < 0.05.

p < 0.01.

***

p < 0.001.

As shown in Table 3, the direct impact of POD on psychological well-being is not statistically significant across lower and higher levels of employment protection. The impact of POD on work–nonwork interference is significant and positive for employees in countries where EPL is relatively low, but for other countries the impact is not statistically significant. JD has a significant negative impact on psychological well-being in countries with relatively low EPL, but not for employees in other countries. The impact of JD on work–nonwork interference is not statistically different across lower and higher contexts of employment protection. We interpret these results to suggest important buffering mechanisms associated with EPL. We argue that the potential threats of POD and JD on different measures of employee well-being are considerably weaker where the national context provides greater employment protection for workers (H2 is thus partially supported).

Table 3 highlights important nuances regarding the buffering role of collective bargaining coverage. POD has a significant negative impact on psychological well-being in countries where bargaining coverage is relatively high, but not where it is relatively low. Unlike EPL, the positive influence of POD on work–nonwork interference is not statistically different across lower and higher contexts of bargaining coverage. Contrary to our prediction in H3, therefore, a higher bargaining coverage may not protect workers against the possible adverse effects of POD on well-being. However, the results involving JD are consistent with H3: it has a significant negative impact on the psychological well-being in countries with a lower bargaining coverage, but not where coverage is relatively high. Similarly, the positive impact of JD on work–nonwork interference is significantly weaker in countries with a higher bargaining coverage. Our prediction in H3 is therefore supported in terms of the results involving JD.

Our analysis of the moderating role of employment contract provided results that are generally consistent with our predictions in H4. As can be seen in Table 3, the negative impact of POD on psychological well-being is significant for employees with a permanent contract, but the impact for those with a non-permanent contract is not statistically significant. In terms of the negative impact of POD on work–nonwork interference, we found no statistically different effects across both types of employment contracts. The results involving JD are generally consistent with our predictions in H4: we found evidence that the negative impact of JD on psychological well-being is statistically significant among employees with a permanent contract, but not for those with a non-permanent contract. The impact of JD on work–nonwork interference is however not significantly different across the two types of employment contracts. In all, H4 is supported for the regression paths involving psychological well-being only.

Table 3 shows that the impact of POD on psychological well-being is significant neither for non-unionists nor for union members. Although the impact of POD on work–nonwork interference is significant and positive, it is not statistically different across non-unionists and union members. We can infer therefore that the moderating effects involving POD deviate from our predictions in H5. Unlike POD, however, the results involving JD provide support for H5: it has a significant negative impact on the psychological well-being of non-unionists but not of union members. Similarly, there is a significant positive impact of JD on work–nonwork interference among non-unionists but not among union members. The findings involving JD are therefore consistent with the argument of Douglas et al. (2017) that union membership offers some protection against the likely adverse effects of job-related insecurities on workers’ well-being.

Discussion

It is generally accepted that the 2008 financial crisis had significant adverse consequences for organizations across Europe, but how the recession that followed this crisis has shaped workers’ experiences of work and well-being is only now being explored systematically (Green et al., 2016). Our study adds to this emerging body of research by focusing on two aspects of recessionary changes at work, POD and JD, and examining their effects on employee well-being. Beyond focusing on these types of recessionary change, our aim was to provide a more institutionally informed perspective on their effects. Thus, we tested the proposition that institutional factors operating at both national and workplace levels will moderate the adverse consequences of recessionary changes on employee well-being. Our starting proposition was that protective institutions would ameliorate these. However, as summarized in Table A3 of the Online Appendix, our results present a more complex story of how institutions influence these relationships.

We found evidence that employees are likely to report a significant decline in the overall quality of their jobs during periods of economic recession. Such changes were in turn associated with diminished levels of psychological well-being and higher levels of work–nonwork interference. Our results, however, showed that the source of these recessionary changes – either organizational (POD) or job-specific (JD) – is important. For example, JD was related to work–nonwork interference, but POD was not. It could be that POD is somewhat remote from an individual’s general life; thus its potential spillover effect from the work to nonwork domain might be weak. In contrast, JD is more likely to be experienced directly by the individual and therefore would have a more direct spillover effect on both general well-being and the work–family interface. Overall, though, our findings are consistent with prior studies, which have generally shown that during economic recession, workers are exposed to work-related stressors that induce feelings of frustration and culminate in ill-being (Houdmont et al., 2012).

Moreover, we found unexpected nuances associated with the four institutional factors examined. National-level institutions designed to protect workers from adverse employment conditions, rather than those operating directly at workplace level, appeared to be associated with stronger buffering effects on employee well-being. We found evidence to support the idea that well-functioning national institutions may provide employees with a sense of employment stability during recessionary periods (Groot et al., 2011). In addition, prior research indicates that the relationship between perceived employment insecurity and well-being may be moderated by a number of other factors, including age, employment status, gender, occupation and family support (Burgard and Kalousova, 2015). Heyes (2011) argued that the adverse consequences were unevenly distributed across the EU workforce, with younger persons having a greater risk of job loss. As our data do not allow us to isolate a wider array of factors as part of our analysis, future research is needed to tap into these complexities further.

In contrast, institutional arrangements at workplace level appeared to reflect the paradoxical patterns we had expected to emerge. Workers employed on permanent work contracts were generally more likely to report adverse experiences of POD and poor psychological well-being than those with less permanent contracts. Similarly, workers with permanent contracts were more likely to report deterioration in job quality than those without. The effects of recessionary events on experienced work–nonwork interference were, however, not statistically different between permanent and non-permanent workers. Unlike employment contract, the moderating effects of union membership were less paradoxical: the adverse effects of JD on both measures of employee well-being were relatively weaker among union members than non-members.

Our study makes three important contributions to the literature on the consequences of the 2008 recession for both organizations and workers. Most significantly, our results highlight critical mechanisms through which recession appears to affect workers, their psychological well-being and experience of work–nonwork interference. The recessionary module included in the 2010 wave of the ESS provides us with a unique opportunity to explore such mechanisms by linking the timing of data collection (immediately following the onset of recession) with individual experiences at work.

The second significant contribution of our study concerns the role of protective institutions in moderating the impact of recession in different countries. Our results suggest that institutions at the national level can have important buffering effects against the adverse health and psychological consequences of recessionary events. This was especially clear for institutions such as EPL that apply to workers across employment levels. However, for workplace-level institutions, such as employment contract and union membership, the overall buffering effects appeared to be less clear-cut.

Our third major contribution relates to our understanding of how labour market institutions work. Our focus is on a particular period in which institutions are ‘activated’ by individuals and organizations during recessionary circumstances. Our results suggest that understanding the impact of institutions requires an in-depth consideration of the conditions under which particular institutional features are likely to have more intensive effects. This proposition has been well established in studies of democratic institutions (Weaver and Rockman, 2010) and the impact of product market and labour market institutions on cross-country trade (Nunn, 2007). Our results suggest a similar effect for labour market institutions and show that they have more intense effects during recessions. One immediate consequence of these results is to highlight the differential or often countervailing effects of institutions, which need to be accounted for in assessing their efficiency and policy implications.

One important gap in our analysis concerns the absence of specific social mechanisms through which institutions might influence employee well-being. Prior research suggests that this may be captured through three pathways: economic factors (such as the absence of savings or other alternative sources of income and consumption), physical health (Baumbach and Gulis, 2014) and socio-psychological pathways (McKee-Ryan et al., 2005) that capture the impact of downturns, periods of risk and uncertainty on individuals’ sense of self-worth and dignity. We also recognize the limitations associated with reliance on cross-sectional data. This precludes us from drawing causal inferences. We have sought to avoid any common method variance problems by matching data drawn from different sources (ESS and OECD), as well as taking advantage of the added module dealing with the effects of the recession administered as part of the 2010 ESS.

Our results point to a number of fruitful directions for future research. First, there is scope for extending our analysis to the potential moderating role of other state-centred and social institutions on the consequences of recession for worker well-being. For example, the politics of social welfare highlights the potential role of family structures and other social networks in alleviating the effects of economic uncertainty (Scruggs and Allan, 2008). Further work also needs to extend this to understand the relative importance of different pathways through which recessionary changes are likely to influence employee well-being, as well as the potentially contradictory effects of different types of institutions. As we have already noted, prior research posits a number of possible mechanisms, but to our knowledge, no study has examined the relative importance of such mechanisms simultaneously. Finally, further work will require longitudinal data to deepen our understanding of how such recessionary effects evolve over time. This may also point to the value of employing qualitative, case-study approaches to fieldwork. Whatever approach is taken, the importance of this topic remains considerable in the light of continuing economic uncertainty and other likely disruptions in the future.

Supplemental Material

Supplementary_Appendix – Supplemental material for Recessionary changes at work and employee well-being: The protective roles of national and workplace institutions

Supplemental material, Supplementary_Appendix for Recessionary changes at work and employee well-being: The protective roles of national and workplace institutions by Chidiebere Ogbonnaya, Peter Gahan and Constanze Eib in European Journal of Industrial Relations

Footnotes

Funding

The author(s) received no financial support for the research, authorship, and/or publication of this article.

Supplemental material

Supplemental material for this article is available online.

Author biographies

Chidiebere Ogbonnaya is a Senior Lecturer in Organisational Behaviour and Human Resource Management in the University of Sussex Business School, UK.

Peter Gahan is a Professor of Human Resource Management and Organisational Behaviour in the Department of Management & Marketing, University of Melbourne, Australia.

Constanze Eib is an Associate Senior Lecturer in the Department of Psychology, Uppsala University, Sweden.

References

Appelbaum

(2011) Macroeconomic policy, labour market institutions and employment outcomes. Work, Employment and Society 25: 596–610.

Baumbach

Gulis

(2014) Impact of financial crisis on selected health outcomes in Europe. European Journal of Public Health 24: 399–403.

Brochu

Morin

(2012) Union membership and perceived job insecurity: Thirty years of evidence from the American General Social Survey. ILR Review 65: 263–285.

Bryson

Freeman

(2013) Employee perceptions of working conditions and the desire for worker representation in Britain and the US. Journal of Labor Research 34: 1–29.

Bryson

Barth

Dale-Olsen

(2013) The effects of organizational change on worker well-being and the moderating role of trade unions. Industrial & Labor Relations Review 66: 989–1011.

Budd

Spencer

(2015) Worker well-being and the importance of work: Bridging the gap. European Journal of Industrial Relations 21: 181–196.

Burgard

Kalousova

(2015) Effects of the Great Recession: Health and well-being. Annual Review of Sociology 141: 181–201.

Byrne

(2012) Structural Equation Modeling with Mplus: Basic Concepts, Applications, and Programming. Mahwah, NJ: Erlbaum.

Campos Lima

Artiles

(2011) Crisis and trade union challenges in Portugal and Spain: Between general strikes and social pacts. Transfer 17: 387–402.

10.

Carr

Chung

(2014) Employment insecurity and life satisfaction: The moderating influence of labour market policies across Europe. Journal of European Social Policy 24: 383–399.

11.

Cascio

(2005) Strategies for Responsible Restructuring. Academy of Management Executive 19: 39–50.

12.

Cheng

Chan

(2008) Who suffers from job insecurity? A meta-analytic review. Applied Psychology 57: 272–303.

13.

Cingano

Leonardi

Messina

, et al. (2010) The effects of employment protection legislation and financial market imperfections on investment: Evidence from a firm-level panel of EU countries. Economic Policy 25: 117–163.

14.

Curl

Kearns

(2015) Financial difficulty and mental wellbeing in an age of austerity: The experience in deprived communities. Social Policy & Society 14: 217–240.

15.

Datta

Guthrie

Basuil

, et al. (2010) Causes and effects of employee downsizing: A review and synthesis. Journal of Management 36: 281–348.

16.

De Cuyper

De Witte

(2007) Job insecurity in temporary versus permanent workers: associations with attitudes, well-being and behaviour. Work and Stress 21: 65–84.

17.

Donnellan

Hanrahan

Hennessy

(2012) Defining an institutional framework for the labour market: Factor Markets Working Paper 24, Brussels: Centre for European Policy Studies. Available at: http://aei.pitt.edu/58536/1/Factor_Markets_24.pdf

18.

Douglas

Haar

Harris

(2017) Job insecurity and job burnout: Does union membership buffer the detrimental effects? New Zealand Journal of Human Resource Management 17: 23–40.

19.

European Social Survey (ESS) (2016) ESS-5 2010 Documentation Report Edition 4.1. Bergen: European Social Survey Data Archive, Norwegian Centre for Research.

20.

Felstead

Green

Jewson

(2012) An analysis of the impact of the 2008–9 recession the provision of training in the UK. Work, Employment and Society 26: 968–986.

21.

Flavin

Pacek

Radcliff

(2010) Labor unions and life satisfaction: Evidence from new data. Social Indicators Research 98: 435–449.

22.

Furåker

Bengtsson

(2013) Collective and individual benefits of trade unions: A multi-level analysis of 21 European countries. Industrial Relations Journal 44: 548–565.

23.

Giorgi

Shoss

Leon-Perez

(2015) Going beyond workplace stressors: Economic crisis and perceived employability in relation to psychological distress and job dissatisfaction. International Journal of Stress Management 22: 137–158.

24.

Green

Felstead

Gallie

, et al. (2016) Job-related well-being through the great recession. Journal of Happiness Studies 17: 389–411.

25.

Green

Mostafa

Parent-Thirion

, et al. (2013) ‘Is job quality becoming more unequal? Industrial and Labor Relations Review 66: 753–784.

26.

Griffith

Macartney

(2014) Employment protection legislation, multinational firms, and innovation. Review of Economics & Statistics 96: 135–150.

27.

Groot

Möhlmann

Garretsen

, et al. (2011) The crisis sensitivity of European countries and regions: stylized facts and spatial heterogeneity. Cambridge Journal of Regions, Economy and Society 4: 437–456.

28.

Heyes

(2011) Flexicurity, employment protection and the jobs crisis. Work, Employment and Society 25: 642–657.

29.

Houdmont

Kerr

Addley

(2012) Psychosocial factors and economic recession: The Stormont study. Occupational Medicine 62: 98–104.

30.

Iverson

Zatzick

(2011) The effects of downsizing on labor productivity: The value of showing consideration for employees’ morale and welfare in high-performance work systems. Human Resource Management 50: 29–44.

31.

Jiang

Probst

Sinclair

(2013) Perceiving and responding to job insecurity: The importance of multilevel contexts. In: Antoniou

ASG

Cooper

(eds) The Psychology of the Recession on the Workplace. Cheltenham: Edward Elgar, pp. 176–95.

32.

Lane

Milesi-Ferretti

(2010) The cross-country incidence of the global crisis. London: CEPR Discussion Paper 7954. Available at: http://citeseerx.ist.psu.edu/viewdoc/download?doi=10.1.1.445.9548&rep=rep1&type=pdf

33.

McKee-Ryan

Song

Wanberg

, et al. (2005) Psychological and physical well-being during unemployment: A meta-analytic study. Journal of Applied Psychology 90: 53–76.

34.

Muthén

(2010) Mplus User’s Guide. Los Angeles, CA: Muthén & Muthén.

35.

Nunn

(2007) Relationship-specificity, incomplete contracts and the pattern of trade. Quarterly Journal of Economics 122: 569–600.

36.

OECD (2016) OECD.Stat (database). Available at: http://stats.oecd.org/

37.

Reeves

Karanikolos

Mackenbach

, et al. (2014) Do employment protection policies reduce the relative disadvantage in the labour market experienced by unhealthy people? A natural experiment created by the Great Recession in Europe. Social Science & Medicine 121: 98–108.

38.

Rigotti

Mohr

Isaksson

(2015) Job insecurity among temporary workers: Looking through the gender lens. Economic and Industrial Democracy 36: 523–547.

39.

Rubery

Rafferty

(2013) Women and recession revisited. Work, Employment and Society 27: 414–432.

40.

Scherbaum

Ferreter

(2009) Estimating statistical power and required sample sizes for organizational research using multilevel modelling. Organizational Research Methods 12: 347–367.

41.

Scruggs

Allan

(2008) Social stratification and welfare regimes for the twenty-first century: Revisiting ‘the three worlds of welfare capitalism. World Politics 60: 642–664.

42.

Shimer

(2010) Labor Markets and Business Cycles. Princeton, N.J: Princeton University Press.

43.

Shoss

(2017) Job insecurity: An integrative review and agenda for future research. Journal of Management 43: 1911–1939.

44.

Snorradóttir

Tómasson

Vilhjálmsson

, et al. (2015) The health and well-being of bankers following downsizing: A comparison of stayers and leavers. Work, Employment and Society 29: 738–756.

45.

Sverke

Hellgren

(2002) The nature of job insecurity: Understanding employment uncertainty on the brink of a new millennium. Applied Psychology 51: 23–42.

46.

Valizade

Ogbonnaya

Tregaskis

, et al. (2016) A mutual gains perspective on workplace partnership: Employee outcomes and the mediating role of the employment relations climate. Human Resource Management Journal 26: 351–368.

47.

van den Bos

Lind

(2002) Uncertainty management by means of fairness judgments. In: Zanna

(ed.) Advances in Experimental Social Psychology. Boston: Elsevier, pp. 1–60.

48.

Warren

(2015) Work-time underemployment and financial hardship: Class inequalities and recession in the UK. Work, Employment and Society 29: 191–212.

49.

Weaver

Rockman

(2010) Why Do Institutions Matter? Washington, DC: Brookings Institution Press.

50.

Wood

Michaelides

(2016) Hindrance and challenge stressors and well-being based work-nonwork interference: A diary study of portfolio workers. Human Relations 69: 111–138.

51.

Wood

Ogbonnaya

(2018) High-involvement management, economic recession, well-being and organizational performance. Journal of Management 44: 3070–3095.

52.

Wood

Daniels

Ogbonnaya

(2018) Use of work-nonwork supports and employee well-being: The mediating roles of job demands, job control, supportive management and work-nonwork conflict. International Journal of Human Resource Management. Epub ahead of print 25 July 2018. DOI: 10.1080/09585192.2017.1423102.

Supplementary Material

Please find the following supplemental material available below.

For Open Access articles published under a Creative Commons License, all supplemental material carries the same license as the article it is associated with.

For non-Open Access articles published, all supplemental material carries a non-exclusive license, and permission requests for re-use of supplemental material or any part of supplemental material shall be sent directly to the copyright owner as specified in the copyright notice associated with the article.

0.00 MB

0.23 MB